- Email: [email protected]
A Model-based Scanning Scheme for CT Metal Artifact Reduction by Constrained Optimization
Proceedings of the 53rd Annual ASTRO Meeting energy X-rays in clinical MV beams are sought. Theoretical studies predict a dose enhancement in MV beams, particularly at depths beyond dmax and for unflattened fields. We quantify the relative in vitro dose enhancement due to the presence of GNPs in a clinical 6 MV beam for conventional and flattening filter free deliveries at various depths. Materials/Methods: HeLa cells were cultured on segmented plates consisting of separate chambers. In two of the chambers, the cells were combined with polymer-coated 50 nm GNPs at a concentration of 0.04 mg/mL. Four chambers contained cells without any nanoparticles. Each plate was surrounded by tissue equivalent material and placed within stacks of solid water. The cells were irradiated by a clinical 6 MV beam at depths of 1.5 cm, 5 cm and 10 cm (SAD setup). Conventional (flattened) and flattening filter free beams were used. Cells were fixed four hours post irradiation and DNA double strand breaks (DSBs) were evaluated by gH2AX staining. Sham plates with and without nanoparticles were also analyzed. The dose enhancement was defined as the ratio of average foci/nucleus in the presence of nanoparticles vs. those without. A two-sample t test is used to disprove the null hypothesis and establish statistical significance. Results: Statistically significant (p \ 0.001) dose enhancement was observed for all depths and delivery modes. Relative to the shallowest depth, dose enhancement was observed as a function of increasing depth beyond 5 cm for the conventional delivery. For flattening filter free delivery, no substantial change in dose enhancement was observed as a function of depth, although the dose enhancement was found to be significantly higher than the conventional delivery at 1.5 cm and 5 cm depths. Cells that received GNPs but no irradiation exhibited only minor increases in DSBs. Conclusions: The flattening filter free delivery caused more DNA DSBs than the conventional delivery at shallow depths. At deeper depths, beam softening of the conventional delivery resulted in similar DNA DSBs as the flattening filter free delivery. Although the in vitro experimental setup is only an approximation of the proposed in vivo scenario of endothelial cell targeting (particularly the minute concentration of GNPs used in this study) the measured relative dose enhancement validates the theoretically predicted trends in dose enhancement as a function of depth and delivery mode for a clinical MV beam. The results of this study open new possibilities for the clinical development of gold nanoparticle aided radiation therapy. Author Disclosure: R.I. Berbeco: None. H. Korideck: None. W. Ngwa: None. S. Johnson: None. B. Price: None. G.M. Makrigiorgos: None.
304
A Model-based Scanning Scheme for CT Metal Artifact Reduction by Constrained Optimization
X. Zhang, L. Xing Stanford University School of Medicine, Stanford, CA Purpose/Objective(s): The streak artifacts caused by metal implants have been a problem that limits various applications of CT imaging, such as target delineation and accurate dose calculation. In the presence of metal shadows, the raw projection data can be viewed as ‘‘incomplete’’. The missing information in the sinogram is where the metal artifacts arise from, and how to deal with the missing data is essential in metal artifact problems. In this work we want to investigate a method that can minimize the missing information and reconstruct images with significantly reduced metal artifacts. Materials/Methods: A penalized-weighted-least-squares method is first used to accurately identify the metal objects in image space. Based on this prior knowledge, a new model-based scanning scheme is designed by shifting the object center during a CBCT scan to avoid X-rays passing through metal regions and reduce the missing projections. In reality, it can be done by moving the patient couch during a CT scan. An iterative algorithm based on constrained optimization is then used for the image reconstruction. It minimizes a quadratic edge-preserving smoothness measure function of the image, subject to the constraint that the estimated projection data is within a specified tolerance of the available metal-shadow-excluded projection data, with image non-negativity enforced. The constrained minimization problem is optimized through the combination of projection onto convex sets (POCS) and steepest gradient descent of the smoothness measure objective. The algorithm is evaluated using a numerical QA phantom (350 x 350 x 16, 1 mm3, only central slice considered) with simulated Poisson noise in the projections. The new scanning scheme is modified over a conventional half-fan scanning geometry with source-to-axis and source-to-detector distances of 100 cm and 150 cm, respectively. Total 339 views projection data are simulated over 360 rotation. Results: Studies showed that the constrained optimization with the model-based scanning data has superior performance compared with analytical FDK reconstruction and conventional iterative reconstructions. It also improved over the constrained optimization with the traditional half-fan scanning geometry. It significantly suppressed metal artifacts in the presence of noise. Profile comparisons and RMSE measurements also suggested that the model-based scanning scheme can effectively reduce the missing information and yield better images. Conclusions: The proposed algorithm can be used to significantly reduce metal artifacts to produce clinically acceptable image for current on-board CBCT image systems. Author Disclosure: X. Zhang: None. L. Xing: None.
305
The Impact of MVCT Imaging Frequency Reduction and Manual Registration Suppression on PTV Margins: A Retrospective Analysis on 259 Patients Treated on Helical Tomotherapy
P. Meyer, S. Chami, E. Enderlin, C. Niederst, D. Jarnet, R. Guerra, D. Karamanoukian, G. Noel Centre Paul Strauss, Strasbourg, France Purpose/Objective(s): It is standard practice that patients treated on helical tomotherapy (HT) have daily MVCT imaging performed before each treatment fraction. These MVCT images are compared to the kVCT planning images, the registration procedure being automatically performed by software (automatic deviations), then manually corrected by the radiation therapists (total deviations). We measured transversal angular (roll), lateral, cranial-caudal and antero-posterior set-up errors of patients treated for different tumor sites to investigate the necessity of the manual registration and the potential reduction of imaging frequency. Materials/Methods: About 6500 shifts of 259 patients treated with HT were recorded and analyzed: 87 patients treated for headand-neck tumors with 9-point fixation thermoplastic masks (TM), 42 and 38 patients treated for abdominal-pelvic tumors
S151
304
A Model-based Scanning Scheme for CT Metal Artifact Reduction by Constrained Optimization
X. Zhang, L. Xing Stanford University School of Medicine, Stanford, CA Purpose/Objective(s): The streak artifacts caused by metal implants have been a problem that limits various applications of CT imaging, such as target delineation and accurate dose calculation. In the presence of metal shadows, the raw projection data can be viewed as ‘‘incomplete’’. The missing information in the sinogram is where the metal artifacts arise from, and how to deal with the missing data is essential in metal artifact problems. In this work we want to investigate a method that can minimize the missing information and reconstruct images with significantly reduced metal artifacts. Materials/Methods: A penalized-weighted-least-squares method is first used to accurately identify the metal objects in image space. Based on this prior knowledge, a new model-based scanning scheme is designed by shifting the object center during a CBCT scan to avoid X-rays passing through metal regions and reduce the missing projections. In reality, it can be done by moving the patient couch during a CT scan. An iterative algorithm based on constrained optimization is then used for the image reconstruction. It minimizes a quadratic edge-preserving smoothness measure function of the image, subject to the constraint that the estimated projection data is within a specified tolerance of the available metal-shadow-excluded projection data, with image non-negativity enforced. The constrained minimization problem is optimized through the combination of projection onto convex sets (POCS) and steepest gradient descent of the smoothness measure objective. The algorithm is evaluated using a numerical QA phantom (350 x 350 x 16, 1 mm3, only central slice considered) with simulated Poisson noise in the projections. The new scanning scheme is modified over a conventional half-fan scanning geometry with source-to-axis and source-to-detector distances of 100 cm and 150 cm, respectively. Total 339 views projection data are simulated over 360 rotation. Results: Studies showed that the constrained optimization with the model-based scanning data has superior performance compared with analytical FDK reconstruction and conventional iterative reconstructions. It also improved over the constrained optimization with the traditional half-fan scanning geometry. It significantly suppressed metal artifacts in the presence of noise. Profile comparisons and RMSE measurements also suggested that the model-based scanning scheme can effectively reduce the missing information and yield better images. Conclusions: The proposed algorithm can be used to significantly reduce metal artifacts to produce clinically acceptable image for current on-board CBCT image systems. Author Disclosure: X. Zhang: None. L. Xing: None.
305
The Impact of MVCT Imaging Frequency Reduction and Manual Registration Suppression on PTV Margins: A Retrospective Analysis on 259 Patients Treated on Helical Tomotherapy
P. Meyer, S. Chami, E. Enderlin, C. Niederst, D. Jarnet, R. Guerra, D. Karamanoukian, G. Noel Centre Paul Strauss, Strasbourg, France Purpose/Objective(s): It is standard practice that patients treated on helical tomotherapy (HT) have daily MVCT imaging performed before each treatment fraction. These MVCT images are compared to the kVCT planning images, the registration procedure being automatically performed by software (automatic deviations), then manually corrected by the radiation therapists (total deviations). We measured transversal angular (roll), lateral, cranial-caudal and antero-posterior set-up errors of patients treated for different tumor sites to investigate the necessity of the manual registration and the potential reduction of imaging frequency. Materials/Methods: About 6500 shifts of 259 patients treated with HT were recorded and analyzed: 87 patients treated for headand-neck tumors with 9-point fixation thermoplastic masks (TM), 42 and 38 patients treated for abdominal-pelvic tumors
S151